Spousal Traumatic Brain Injury and Risk of Mortality, Dementia, and Cardiovascular Events

Importance Traumatic brain injury (TBI) is highly prevalent in older adults. While there is strong and growing evidence that TBI is associated with caregiver burden and psychological health, it remains unclear whether a spouse’s TBI is associated with the long-term health of the uninjured partner. Objective To evaluate whether having a spouse who sustains an incident TBI is associated with risk of mortality, dementia, and major adverse cardiovascular events (MACE) and to assess whether associations differ by sex or survival of the injured spouse after TBI. Design, Setting, and Participants This prospective community-based cohort of married couples was performed at 4 US centers in North Carolina, Mississippi, Minnesota, and Maryland as part of the Atherosclerosis Risk in Communities (ARIC) Study. Of 4506 baseline spousal pairs, 3281 were retained for mortality and dementia analyses and 2740 of these for MACE analyses. Participants contributed person-time from baseline (1987-1989) until incident outcome, death, loss to follow-up, divorce or separation, or administrative censoring on December 31, 2022. Analyses were conducted from November 24, 2025, through August 25, 2026. Exposure Incident spousal TBI (ascertained using self-report, hospitalization surveillance, and linked Centers for Medicare & Medicaid Services data), modeled as a 2-category time-varying exposure (spousal TBI vs no spousal TBI) and 4-category exposure incorporating spousal death. Main Outcomes and Measures Incident all-cause mortality, dementia, and MACE. Cox proportional hazards regression shared frailty models were used to estimate associations adjusted for demographic, clinical, and behavioral characteristics. Results Among 3281 married pairs included in mortality and dementia analyses (among whom, 2740 pairs were included in MACE analyses), the mean (SD) age at baseline was 54.5 (5.5) years, with 3281 female and 3281 male spouses (for a total of 6562 participants). The median follow-up was 19.5 (IQR, 10.6-29.0) years. Spousal TBI was associated with increased risk of mortality (hazard ratio [HR], 1.24 [95% CI, 1.07-1.43]; P = .004) but not dementia (HR, 1.17 [95% CI, 0.88-1.56]; P = .28) or MACE (HR, 1.15 [95% CI, 0.84-1.57]; P = .40). In sex-stratified analyses, associations were similar among males and females. In models incorporating spousal death, mortality risk was higher among individuals whose spouse was alive with a TBI (HR, 1.22 [95% CI, 1.05-1.41]; P = .008), higher among those whose spouse was deceased without a TBI (HR, 1.48 [95% CI, 1.36-1.61]; P < .001), and higher among those whose spouse was deceased with a TBI (HR, 1.27 [95% CI, 1.07-1.52]; P = .006) compared with those whose spouse was alive without a TBI. Results were similar, but less precise, for dementia and MACE. Conclusions and Relevance In this cohort study of married pairs of participants in the ARIC Study, having a spouse who sustained a TBI was associated with a modest but significantly higher risk of mortality in the uninjured partner, whereas findings for dementia and major adverse cardiovascular events were less consistent. These findings suggest that the consequences of TBI may extend beyond injured individuals and support family- and care partner–centered approaches to care.